Research On Fractal Characteristics Of Low-voltage Power Line Communication Channel

With its violent time-varying, strong interference and severe attenuation, the analyses of this communication channel become a challenging but important basic subjects for the sustainable development of L-PLC (Low-Voltage Power Line Communication) technology. In this study, we point to bring fractal theory to the characterization and perdiction of L-PLC channel, this papre can be divided into two main parts: the determination of fractal characteristics and the fractal prediction of this channel, detail works are as follows:1. The time series measured from L-PLC network are analyzed by R/S method which provides us a quantitative analysis of long-term time-dependence of the time series. The results has showed: Hurst index of measured time series are greater than 0.5, fractal dimension is greater than 1.0, L-PLC channel has obvious fractal and chaotic characteristics.2. The changing curves of power spectrum and statistical moment of these time series are drawn, for the analyzing of multi-fractal characteristics of this channel. The results has showed: statistical moment is a nonlinear function, and multi-fractal spectrum is strict symmetrical, the L-PLC channel has strong multi-fractal properties.3. In the end, a new fractal prediction algorithm is proposed basing on the weak chaotic and statistical self-similarity of the L-PLC channel signal. In the algorithm, the fractal interpolation based on phase space reconstruction sample selection method and fractional collage theory is applied to construct an iterated function system, and then set up the fractal prediction model. The effectiveness of the algorithm is checked with actual noise signals. The results show that, the average prediction error is smaller than 6.0%, prediction accuracy is improved about 20% comparing with traditional algorithm.